BeCu spring deformation

[RayJohnson2]

Hi all,

Could you help refresh my memory a bit?
I am designing a 'contact needle' in beryllium copper (BeCu).
The purpose of this needle is to make contact with the lead of an electronic semiconductor device (a 'chip') to do electric measurements on this chip.
To make contact, the needle is pressed against the lead of the chip.
By doing this, the needle is compressed and it acts like a spring. But the needle also (temporarily) deforms and internal stresses are created.
It is of course vital that, when the contact is removed, that the needle returns to its original shape.

I have been doing some FEA simulations to get an impression of the deformed shape and to get an idea of the stresses that occur.
Von Mises stress seems the most appropriate.

Now my question: If the maximum Von Mises stress that occurs is smaller than the Tensile strength of the material, can I then conclude that the material will not undergo plastic deformation? In other words, that there is no permanent deformation and that the needle will come back to its original shape.

(image attached, and another one in my next post)

Kind regards

 

[RayJohnson2]

Here is another image of the simulation.

 

[dgallup]

You want to stay under the yield strength of the material, not the tensile strength. You also probably need to take fatigue into account if this needle is going to be used more than just a few times.

----------------------------------------

The Help for this program was created in Windows Help format, which depends on a feature that isn't included in this version of Windows.

 

[EdStainless]

And if you can get a tensile curve look for where the load line starts deviating from straight.
At the engineering yield stress you already have 0.2% deformation, not a good thing to do over and over.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[winstonsk]

I can't imagine loading a beryllium copper pin enough that the pin would deform before cracking the chip.
Those alloys can be as strong as tool steel.

 

[RayJohnson2]

@dgallup: Ok, I will look at yield strength.
Fatigue will be less of an issue. This setup will be used for burn-in tests of chips. One test lasts multiple days, in an oven (furnace) .. So there will not be a huge number of compression-decompression cycles of the needle.

@EdStainless: Yes, that 0.2% remaining deformation is already too much. I agree.

@winstonsk: the pin is not directly pushing onto the chip (the silicon die), but rather on the leads of the packaged chip. So that is a metal-to-metal contact.
I have attached a picture of the (packaged) chip. The needles are pushing on the metal pins.

To get a feeling of dimensions: the distance between each of the 8 pins of the chip in the picture is 1.27 mm (0.05 inches).
The total length of the needle (see pics in my first 2 posts) is about 6 mm (0.236 inches).

 

[RayJohnson2]

The beryllium copper used is UNS C17200 ( DIN 17666, CW101C, werkstoffnr 2.1247).
This is available in different hardnesses. The supplier describes this version as 'half hard'.
Do you think that the hardness has a big influence on the yield strength and/or Young's Modulus ?

 

[EdStainless]

The modulus does not change, but the strength does.
You need to design these with a cantilever or flex portion so that you can control where they flex and by how much.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[RayJohnson2]

@EdStainless: I am changing the design, checking deformation and Von Mises stress in LISA, a finite element software.
But the numbers that I enter must be correct, of course .. "Garbage in = garbage out" ..